Children whose parents divorce are often at significantly increased risk for a number of mental health problems, including childhood depression, conduct disorder, and school problems. The present application proposes the development and evaluation of a preventive intervention to ameliorate this risk targeted at the recently divorced non-custodial father (NCF), who is known to have profound effects on the mental health outcomes of his child. We reformulate findings about the mechanisms of these effects into a theory of the mediators of the NCF's impact on the child. This theory then drives the design of a preventive intervention which should theoretically benefit the mental health of the child. The theory focuses on the two key mediators of the child's mental health within the NCF's control: Interparental conflict, and the quality of the NCF-child relationship. In turn, these two mediators should be impacted by: (1) the NCF's perceived control over post-divorce family issues, (2) his motivation and skills for effectively managing conflict with the ex- spouse, (3) his level of commitment to the parenting role, and (4) his level of basic parenting skills, variables which become the immediate change goals of our 10-session group intervention, DADS FOR LIFE. In two simultaneous experimental field trials, we will evaluate a total sample of 320 recently divorced families in which the NCF is randomly assigned to either receive this program or a placebo (literature only) control. The first trial involves men whose ex-wives have chosen to simultaneously participate in New Beginnings. our efficacious program for custodial mothers (CMs), and will assess the additive effects of DADS FOR LIFE; the second trial involves men whose wives declined participation in the mother program and assesses the impact of our program when taken alone. The participants are expected to be approximately equally distributed into the two experiments. Each is a 2 X 2 X 2 factorial. Gender of target child will be a second factor in each design. Age of target child is the final factor, younger (4-7) vs older (8-12). The evaluation features multiple measure, multiple informant assessment of families, on outcome and putative mediator variables at Pre-test, Post- test and a one year Follow-up. Interviews with both parents are by telephone. Children's reports, and teacher report questionnaires of the major dependent variable, child's mental health, will also be obtained at all three waves. Analyses of covariance, measurement modeling techniques, and mediational analyses via structural equations models test the impact of the program, and assess the validity of the causal sequence posited in the theory of the intervention. The project also contains two special components: (1) To gauge the program's effect on certain dependent variables requires systematic observation of dyadic interaction, both between NCF and child, and between NCF and CM. Since this work is very intensive, a subsample will participate in these micro-analytic studies of dyadic interaction. (2) The evaluation also assesses the impact of the program on four outcome variables likely to additionally show effects of program impact: mental health of the CM; of the NCF; either spouse's resort to legal actions resulting in Court filings; and child support compliance.